INDEX

Explanations

mentions of different types of wards in a medical or institutional context

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

ward

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 surla

-0.54

Tikang

-0.52

 sacri

-0.47

antaranya

-0.45

={

-0.45

 SDLK

-0.44

Enllaces

-0.44

citrus

-0.44

 Selena

-0.43

 Tsing

-0.43

POSITIVE LOGITS

 ward

1.99

 wards

1.74

 Ward

1.59

Ward

1.52

 WARD

1.50

ward

0.99

病房

0.90

WARD

0.85

wards

0.83

 warden

0.71

Activations Density 0.004%

mentions of different types of wards in a medical or institutional context

ward

No Known Activations